Sei sulla pagina 1di 7

Kasetsart J. (Nat. Sci.

) 46 : 298 - 304 (2012)

Energy Utilization Index and Benchmarking


for a Government Hospital

Sirinath Thongkaw and Kriengkrai Assawamartbunlue*

ABTRACT

The energy utilization index is a measure that can be used to identify whether energy is being
efficiently used. Energy consumption in public buildings is extremely large and is difficult to control and
monitor. An efficient index based on engineering principles and ease of use was developed to monitor
the energy usage in a public building,a hospitalas a prototype building. The index was calculated
as the difference between the actual energy consumption and optimal energy consumption expressed
as a ratio of the optimal energy consumption. Actual energy consumption was defined as the amount
indicated on an energy invoice. Optimal energy consumption was derived based on several engineering
standards and codes as a function of variables such as the number of officers, number of customers,
operating hours and building areas, among other factors. The EnergyPlus simulation program was used
to analyze optimal energy consumption. Based on the results, the prototype hospital could save up to
15% on energy consumption if the building were improved to comply with standards and codes.
Keywords: EnergyPlus, energy utilization index, benchmarking, hospital building, energy standards
and codes

INTRODUCTION BACKGROUND

The current energy crisis has forced both Sharp (1996) analyzed the energy
the government and the private sector to determine intensities of office buildings and used a linear
energy conservation measures in order to lower regression model to identify the strongest
their energy consumption. However, they are first determinants of different intensities of office
confronted with the same simple questions: Do building energy use. The regression model showed
we use energy efficiently nowadays? and How that the most common and strongest indicators
do we know whether we use energy efficiently? of the intensity of electricity use in a building
The answers to such questions lead to different corresponded to the logarithm of the number of
actions. An approach was used to develop an workers per area, the category describing the
energy utilization index, using a hospital as a number of personal computers in the building,
prototype building. the number of operating hours, the number of
occupants, economizers and chillers. Murray et al.
(2008) found that a benchmark of 0.2 GJ.m-3 was
suitable for Scottish healthcare buildings. Energy

Energy Technology Research Group, Department of Mechanical Engineering, Kasetsart University, Bangkok 10900, Thailand.
* Corresponding author, e-mail: fengkka@ku.ac.th

Received date : 06/08/11 Accepted date : 15/03/12


Kasetsart J. (Nat. Sci.) 46(2) 299

Policy and Planning (2007) developed an energy actual energy consumption (AEC) is greater than
index for nine public buildings. Each building had the optimal energy consumption (OEC), the EUI
different equations; for example, the equation for is negative and the building does not efficiently
a hospital division was a function of occupancy use energy. On the other hand, if the actual energy
number, total usage area, number of outpatients, consumption (AEC) is less than the optimal energy
length-of-stay of inpatients admitted into the consumption (OEC), the EUI is positive and the
hospital and ambient temperature. Department of building uses energy efficiently.
Alternative Energy Development and Efficiency
(2008) also developed energy indices for several MATERIALS AND METHODS
other industries. For the steel industry, the energy
index was 728 kWh.t-1. The EUI was chosen for this research
The work cited in the previous paragraph because of its flexibility and the variety of factors
provides two common types of energy index. that affect the energy consumption of the selected
One is defined in terms of the specific energy building. EUI is defined by Equation 1:
consumption (SEC), while the other uses an energy
OEC AEC
utilization index (EUI). SEC is the ratio of the EUI = OEC
(1)
energy consumption divided by production. SEC
is commonly used in industries utilizing products where AEC is the actual energy consumption
such as steel, wood and ceramics. The EUI is the indicated on an energy invoice and OEC is
ratio of actual energy consumption divided by the predicted optimal energy consumption
optimal energy consumption. The actual energy based on several engineering standards, laws
consumption is the real energy consumption or recommendations developed by well-known
as indicated on an energy invoice. The EUI is energy institutes such as Thailands Energy
commonly used, with examples being hospitals, Conservation Promotion Act (Ministry of Energy,
schools and commercial buildings. 2007), the Project Performance Improvement
Benchmarking measures and compares of Electrical Equipment (Electricity Generating
the products and services of an individual Authority of Thailand, 2005) and the Energy Star
organization with general corporate practices based Project (U.S. Environmental Protection Agency
on competitors in order to improve the products and the U.S. Department of Energy, 1995).
and services of the individual organization (Robear, A difficulty in calculating the EUI is
2000). Benchmarking is a popular tool nowadays determining the OEC, which depends on several
used by many organizations because they can variables such as the number of staff, number of
make substantial improvements and continue to customers, number of operating hours and the size
expand. Generally, benchmarking is undertaken of air-conditioned areas, among others. The OEC
within the same or a similar industry with regard to should comply with several engineering standards
products or services, such as steel, semiconductors, and laws as indicated above. EnergyPlus (U.S.
hotels, among others. One popular benchmark Department of Energy, 2011) was used as a
tool is to compare the SEC or the EUI. Although simulation tool to predict the optimal energy
benchmarking is very useful, a disadvantage can consumption. The program was first calibrated
be any differences in the corporate culture, so that with the monthly actual energy consumption
this method is not suitable for every organization. for 12 mth, based on building information and
In the current study, the EUI was used to compare Thailand weather data (U.S. Department of
the benchmark performance of the building. If the Energy, 2011). Several parameters were adjusted
300 Kasetsart J. (Nat. Sci.) 46(2)

so that the difference between the monthly actual


and monthly predicted consumption was less than
10%.
After calibration, standard criteria were
applied to predict the monthly optimal energy
consumption if the building had been modified to
meet minimum energy requirements according to
the applicable codes. Then EnergyPlus was used
to generate sets of data by varying factors, such as
occupancy, outdoor temperature and other factors.
These data were used to develop a regression
model that could easily be used without requiring
technical skills. Figure 1 shows the procedure
in the study to determine the optimal energy
consumption for a building.
This approach was chosen because
EnergyPlus is a sophisticated program that
requires engineering skill and experience. It is
not suitable for building technicians to directly
implement the program if there are continual
changes in the operating conditions, such as
the number of outpatients and inpatients and
the ambient temperature, among others. Thus, a
regression model was developed using multi-linear
regression techniques (Equation 2):

Y = B0 + B1 x1 + B2 x2 + + Bk xk (2)

where the constants, B0,,Bk, of the equation


are determined using the basic principles of
statistics and multi-linear regression, given the
set of data generated by EnergyPlus. If the
equation has a high R2 value, the equation can Figure 1 Research approach used in the study.
explain the relationship between the independent
and dependent variables (Vanichbuncha, 2005). PROTOTYPE BUILDING
Hence, the OEC was determined and substituted
into Equation 1 to determine whether or not the A 21-story hospital building with a
building efficiently used energy. If the EUI is total usable area of 55,836 m2 was used as the
positive, the building consumes less energy than prototype building. The air-conditioned area was
the standard. However, if the EUI is negative, approximately 51% of the total area. The walls and
the building consumes more energy than the roof were made of common brick and cement and
standard. covered with a light-colored paint. Fenestrations
were are made of 3mm-thick glass with a thermal
Kasetsart J. (Nat. Sci.) 46(2) 301

barrier. The fenestrations accounted for 44% of in EnergyPlus to predict the optimal energy
the total exterior area. There were several types consumption under various operating conditions.
of rooms in the building; for example outpatient The data from the predictions were then used to
and inpatient departments, examination rooms, develop a regression model. The energy standards
meeting rooms and offices. The prototype building and codes are shown in Table 1.
was modeled by EnergyPlus. There were 526 The regression model was developed
spilt-type air-conditioners, with an average energy to predict the OEC and was based on statistical
efficiency rating (EER) lower than 7.6 BTU. analysis. Four variables were considerednamely,
hr-1.W-1 and 7 air-cooled chillers in the building. the number of outpatients, inpatients, staff and the
Typical lighting consisted of 18 W and 36 W ambient temperature. These variables substantially
fluorescent lamps with electromagnetic ballasts. affect the energy consumption of the building and
Like other hospitals, there was a large amount of have normally been recorded every month by
medical equipment, some of which required large building staff; thus, it was easy for them to retrieve
amounts of power, such as X-ray machines and the these data for use in the developed model. A dataset
blood freezer. for the regression was generated by varying these
four variables using EnergyPlus to represent
RESULTS AND DISCUSSION various operating conditions. The regression
model is shown in Equation 3 (R2 = 0.853):
Figure 2 illustrates the differences (less
than 10%) between the actual monthly energy OEC = exp[8.114 + 0.258 ln(outpatient)
consumption from the electricity bills and that + 0.486 ln(inpatient) 0.404 ln(staff) + 0.552
predicted by EnergyPlus after calibration. The ln(ambient temperature)] (3)
energy standards and codes were implemented

500,000
Predicted monthly energy consumption (kWh.mth-1)

450,000

400,000

350,000

300,000

250,000

200,000

150,000

100,000

50,000

0
0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 450,000 500,000

Actual monthly energy consumption (kWh.mth-1)

Figure 2 Comparison between actual energy consumption and predicted consumption using the
calibrated EnergyPlus model.
302 Kasetsart J. (Nat. Sci.) 46(2)

where outpatient, inpatient and staff are parameters Figure 4 shows the monthly energy
for the total monthly numbers of outpatients, utilization index of the prototype building in 2008.
inpatients and staff, respectively, and ambient Actual data (numbers of outpatients, inpatients,
temperature is the monthly average outdoor employees and ambient temperature) were
temperature. retrieved from the hospital records and substituted
Figure 3 shows the predictions from into Equation 3 to predict the OEC. Then, the OEC
EnergyPlus and from the regression model in and AEC from energy bills were substituted into
Equation 3. Equation 1 to calculate the EUI.

Table 1 Implemented energy standards and codes.


Specification Organization Data source Value
Efficiency of air Ministry of Performance Improvement of 11 Btu.W-1
conditioning Energy Electrical Equipment project
OTTV Ministry of The Energy Conservation 30 W.m-2
Energy Promotion Act (2007)
RTTV Ministry of The Energy Conservation 10 W.m-2
Energy Promotion Act (2007)
Power of lighting Ministry of The Energy Conservation 12 W.m-2
equipment Energy Promotion Act (2007)
OTTV = Overall thermal transfer value; RTTV = Roof thermal transfer value.

500,000

450,000
Prediction from Equation 3 (kWh.mth-1)

400,000

350,000

300,000

250,000

200,000

150,000

100,000

50,000

0
0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 450,000 500,000

Predictions from EnergyPlus (kWh.mth-1)

Figure 3 Predicted optimal energy consumption (OEC) from Equation 3 and the EnergyPlus
software.
Kasetsart J. (Nat. Sci.) 46(2) 303

The monthly EUI was negative because According to the Energy Conservation
approximately 74% of the existing equipment Promotion Act (Ministry of Energy, 2007), all
and air conditioners were classified as low- commercial and industrial buildings must have
performance based on field measurements. The an energy management team to manage the
existing overall thermal transfer value (OTTV) and use of energy within the facility. The EUI can
roof thermal transfer value (RTTV) of the building be used as a performance index for the energy
were 51.67 and 21.49 W.m-2, respectively, which management team as well as for their energy
are above the energy standards shown in Table 1. conservation measures. Required data could be
The prototype hospital could save approximately recorded and used to calculate the EUI to assess
15% on energy consumption costs if the building their performance each month.
were improved to conform to the standards. The The method can also be used for economic
OEC was derived from the energy standards and analysis. If the prototype building were improved
codes already mentioned. To have a positive EUI, to the energy standards, the building would
the equipment and building must be improved to consume less energy (approximately 457,457
meet the energy standards and codes; for example, kWh.yr-1) with an estimated value of THB 1.6
by installing exterior or interior shading to reduce million annually. This amount is simply calculated
the solar heat load or by replacing existing clear from the difference between the actual energy
glass with double low E glass. Such changes would consumption and the optimal energy consumption.
contribute greatly to a reduction in the OTTV but While this estimate may not be accurate, it is
would require a large capital investment. Roof useful for a preliminary assessment and economic
insulation could be installed to reduce thermal analysis of the measures which can be presented
transfer through the roof and reduce the RTTV. to the board of directors. Although it may require
Air conditioners could be changed to ones with a time and money to improve a buildings energy
higher performance rating (EER>11) and office usage, the energy management team could use the
appliances could be replaced with those meeting EUI to develop a plan for implementing energy
the Energy Star standards. saving measures.

Figure 4 Energy utilization index (EUI) analysis by month for the prototype hospital.
304 Kasetsart J. (Nat. Sci.) 46(2)

CONCLUSION Energy Policy and Planning. 2007. Energy index


and benchmarking for government agency.
The energy utilization index is a measure Proceedings of Hearing About the Criteria
that can be used to identify whether energy is for Energy Management Government.
being efficiently used, compared with energy 1819 May 2007. Bangkok,. Thailand.
standards and codes. It also provides a means Ministry of Energy. 2007. The Energy
to estimate the monetary value of energy costs Conservation Promotion Act. 4 December
and potential saving that the energy management 2007. 124(87) : 110.
team can use to make investment decisions and Murray, J., O. Pahl and S. Burek. 2008.
facilitate retrofitting work. The approach proposed Benchmarking NHS Scotlands smaller
to determine the EUI of a building was based on health buildings. Evaluating the Scope for
energy simulation and energy codes. The approach Energy-Efficiency Improvements in the
has limitations which may cause prediction Public Sector. 36: 12361242.
errors such as where a building is modified, for Vanichbuncha, K. 2005. Data Analysis with SPSS
example, by increasing the air-conditioned area, for Windows. Thammasarn Press. Bangkok.
increasing the amount of equipment or changing 520 pp.
the function of an area. Such limitations are due to Robear, P.J. 2000. Benchmarking: A System
the simplicity of the regression model which was Approach for Continual Improvement.
designed for ease of use by non-technical persons. Dhurakij Pundit. Bangkok. 124 pp.
When a building is modified, a new regression Sharp, T. 1996. Energy benchmarking in
model must be developed that can accurately commercial office buildings. Proceedings of
predict the optimal energy consumption under the ACEEE 1996 Summer Study on Energy
the new operating conditions. Thus, Equation 3 Efficiency in Buildings. 4: 321329.
may need to be revised every 35 years, or when U.S. Department of Energy. 2011. EnergyPlus
the function of the building or the amount of Energy Simulation Software. [Available
equipment changes noticeably. from: http://apps1.eere.energy.gov/buildings/
energyplus/]. [Sourced: 5 April 2012].
LITERATURE CITED U.S. Environmental Protection Agency and the
U.S. Department of Energy. 1995. Energy
Department of Alternative Energy Development Star Qualified Products. [Available from:
and Efficiency. 2008. Publication Study http://www.energystar.gov]. [Sourced: 10
Criteria for Energy Steel Industry. Bangkok December 2011].
Press. 52 pp.
Electricity Generating Authority of Thailand. 2005.
The Project Performance Improvement of
Electrical Equipment. [Available from:
http://www2.egat.co.th/labelNo5/index.htm].
[Sourced: 5 December 2010].